Olaparib Pivotal mCRPC Data Published in NEJM as FDA Weighs Approval

Johann De Bono, MB CHB, PhD, MSC

Findings from the pivotal phase 3 PROfound trial of olaparib (Lynparza) as a treatment for men with metastatic castration-resistant prostate cancer (mCRPC) harboring homologous recombination repair (HRR) gene alterations were published in the New England Journal of Medicine,¹ as the FDA considers a supplemental new drug application (sNDA) for the PARP inhibitor in this setting.

The published results, on which the sNDA were based, showed that olaparib led to a 66% reduction in the risk of disease progression or death compared with abiraterone acetate (Zytiga) or enzalutamide (Xtandi; HR, 0.34; 95% CI, 0.25-0.47; P <.001) in patients with BRCA1/2- or ATM-mutant mCRPC.

Under the Prescription Drug User Fee Act, the FDA is scheduled to make a decision on the sNDA in the second quarter of 2020.

“Our findings validate phase 1 and 2 data on the antitumor activity of olaparib in metastatic castration-resistant prostate cancer,” first author Johann de Bono, MB, ChB, PhD, Institute of Cancer Research and Royal Marsden Hospital, and coauthors wrote.

AstraZeneca and Merck (MSD), the codevelopers of olaparib, recently announced that updated data from the PROfound trial showed the PARP inhibitor also significantly improved overall survival (OS) versus abiraterone acetate and enzalutamide.² The companies intend to present the OS data at an upcoming medical meeting.

In the prospective, multicenter, randomized, open-label, phase 3 PROfound trial, investigators evaluated the efficacy and safety of olaparib versus abiraterone or enzalutamide in patients with mCRPC who have progressed on prior treatment with new hormonal anticancer treatments, and also have a qualifying tumor mutation in 1 of 15 genes involved in the HRR pathway, including BRCA1/2, ATM and CDK12.

There were 2 cohorts in the study. In cohort A, patients had alterations in BRCA1/2 or ATM (n = 245), which are more established markers of HRR. In cohort B (n = 142), patients had an alteration in BARD1, BRIP1, CDK12, CHEK1/2, FANCL, PALB2, PPP2R2A, RAD51B/C/D, or RAD54L, which are all associated with HRR but not as established as those in cohort A. Within each cohort, patients were randomized 2:1 to olaparib or physician's choice of abiraterone plus prednisone or enzalutamide.

In cohort A, 162 patients received olaparib and 83 got abiraterone plus prednisone or enzalutamide. In cohort B, 94 patients received olaparib and 48 got physician's choice of abiraterone plus prednisone or enzalutamide. In addition to each cohort, a combined assessment of patients was also analyzed (N = 387). Across cohorts, olaparib was given at 300 mg twice daily, abiraterone at 1000 mg per day with prednisone at 5 mg twice daily, and enzalutamide at 160 mg daily.

Patients characteristics were well-balanced between arms in each group. In cohort A for the olaparib arm, the median age was 68 years (range, 47-86), and 23.5% of patients had metastatic disease as an initial diagnosis. The median baseline prostate-specific antigen (PSA) level was 62.2 ug/L. In both arms, between 42% and 48.2% of patients had received prior enzalutamide and between 38.3% and 34.9% had received prior abiraterone. Both treatments were received by 16.9% to 19.8% of patients. Two-thirds of patients had also received a prior taxane prior to study entry.

In the combined cohorts, the median age in both arms was 69 years and approximately one-quarter of patients had de novo metastatic disease. The median PSA was 68.2 ug/L in the olaparib group and 106.5 ug/L for physician's choice. Patients in the olaparib and physician's choice arms, respectively, received prior enzalutamide (41% and 41.2%), abiraterone (39.1% and 41.2%), both agents (19.9% and 17.6%), and a taxane (66.4% and 64.1%).

In cohort A, the confirmed ORR was 33.3% with olaparib compared with 2.3% with the hormonal therapies (odds ratio [OR], 20.86; 95% CI, 4.18-379.18; P <.0001). The median time to pain progression was not yet reached with olaparib compared with 9.92 months for the hormonal agents, representing a 56% reduction in the risk of pain progression (HR, 0.44; 95% CI, 0.22-0.91; P = .0192). In this cohort, the median radiographic progression-free survival (rPFS) was 7.4 months and 3.6 months with olaparib and hormonal therapy, respectively.

By blinded review, the hazard ratio for rPFS in cohort B was 0.88 (95% CI, 0.58-1.36).

In cohorts A and B combined, the median rPFS with olaparib was 5.8 months and 3.5 months for abiraterone or enzalutamide. The confirmed ORR was 21.7% and 4.5% with olaparib and abiraterone/enzalutamide, respectively (OR, 5.93; 95% CI, 2.01-25.40; P = .0006). The median OS in this group was 17.5 months with olaparib and 14.3 months with abiraterone or enzalutamide (HR, 0.67; 95% CI, 0.49-0.93; P = .0063)

Regarding safety, the tolerability of olaparib was consistent with what has been previously observed in prior studies. The most common adverse events (AEs), occurring at ≥20%, were anemia (47%), nausea (41%), fatigue/asthenia (41%), decreased appetite (30%), and diarrhea (21%). The most common grade ≥3 AEs, occurring at ≥1%, were anemia (22%), fatigue/asthenia (3%), vomiting (2%), dyspnea (2%), urinary tract infection (2%), pulmonary embolism (2%), decreased appetite (1%), diarrhea (1%), back pain (1%), and nausea (%). A total of 18% of patients who were treated with olaparib discontinued treatment due to AEs.

References

1. de Bono J, Mateo J, Fizazi J, et al. Olaparib for Metastatic Castration-Resistant Prostate Cancer [published online April 28, 2020]. N Engl J Med. doi: 10.1056/NEJMoa1911440
2. Lynparza demonstrated overall survival benefit in Phase III PROfound trial for BRCA1/2 or ATM-mutated metastatic castration-resistant prostate cancer. Posted April 24, 2020. https://bit.ly/3bBICv8. Accessed April 24, 2020.

Overall, the trial showed that olaparib demonstrated a 51% reduction in the risk of disease progression or death versus either of the antiandrogen agents (HR, 0.49; 95% CI, 0.38-0.63; P <.001) in the entire population of patients with HRR-mutant mCRPC who had mutations in genes for BRCA1/2, ATM, CDK12, or 11 other HRR-mutated genes.